CONTENT-BASED INFORMATION RETRIEVAL (CBIR)


By using this CBIR method, it is actually the easiest way to get the information that the users wish to search in the fastest speed without being directed to some unrelated content that the users doesn't need. In other words, the result of the query can be shown with more relevant selections first followed by less relevant selections. With this content based retrieval method, which enable users to search multimedia information in terms of the actual content which include image, audio, video, shapes, textures, or any other information that can be derived from the image itself . In short , it is a system that can filter images based in their content and provide a better indexing and give a more accurate results .


WHY CONTENT-BASED IMAGE RETRIEVAL IS NEEDED

When we want to search information in search engines such as Google and others search engines,our search results usually end up with any words that we type into those search engines even though those results don't related to our search.As a result, a lot of unnecessary links or pages that are not related to our search were produced.CBIR filter search results based on their contents would help narrowed down our results to more specific ones."Content-based" means that the search will analyze the actual contents of the image rather than metadata that was use in search engines such as keywords,tags,descriptions and others."Content" might refer to colour,shapes,textures or any other information that can be derived from the image itself.Keywords for images manually enter by humans in a large database can be inefficient,expensive and may not capture every keywords describes the images.

CONTENT-BASED IMAGE RETRIEVAL


It will store info about the locations of the faces in photos. So, the identities of individual appear in photos are important aspect to the system. Retrieval is performed by comparing features of a query images with corresponding features of image stored in database and representing the user with the images that have the most similar features. If the users is not satisfied with the retrieved result, they can research the result by selecting the most relevant to the search query and get the newest information. Besides, this will also help users with different cultural background and language to effectively search the information the want using this method. This system also cluster the image within a single event based on face arrangement.But, the difficulty in the current computer technology still mainly focus on lower level features, such as shape, text and colour.




cbir-sys.GIF


List of CBIR engines

Name
Description
External Image Query
Metadata Query
Index Size (Estimate, Millions of Images)
Organization Type
License (Open/Closed)

Bing Image Search
Microsoft's CBIR engine
No
Yes

Public Company
Closed
Elastic Vision
Smart image searcher with content-based clustering in a visual network.
No
No

Private Company
Closed
Google Image Search
Google's CBIR system, note: does not work on all images
No
Yes

Public Company
Closed
Imense Image Search Portal
CBIR search engine, by Imense.
No
Yes
3M
Private Company
Closed
Imprezzeo Image Search
CBIR search engine, by Imprezzeo.
No
Yes

Private Company
Closed
vSearch Visual Image Search
CBIR search engine, by pixolution
Yes
No
10M
Private Company
Closed
Incogna Image Search
CBIR search engine, by Incogna Inc.
No
Yes
100M
Private Company
Closed
Like.com
Shopping & fashion based CBIR engine
No
Yes
1M
Private Company
Closed
MiPai similarity search engine
Online similarity search engine
Yes
Yes
100M
Individual
Closed
Piximilar
Demo engine, developed by Idee Inc.
No
No
3M
Private Company
Closed
Empora
Product comparison & shopping using CBIR for product images. Previously known as Pixsta
No
Yes
0.5M
Private Company
Closed
Shopachu
Shopping & fashion CBIR engine, by Incogna Inc.
No
Yes
1M
Private Company
Closed
TinEye
CBIR site for finding variations of web images, by Idee Inc.
Yes
No
1800M
Private Company
Closed
Tiltomo
CBIR system using Flickr photos
No
Yes

Private Company
Closed
eBay Image Search
Image Search for eBay Fashion
No
Yes
20M
Public Company
Closed

CBIR research projects/demos/open source projects

Name
Description
External Image Query
Metadata Query
Index Size (Estimate, Millions of Images)
Organization Type
License (Open/Closed)
ALIPR
Developed by Penn State University researchers
Yes
Yes

University
Closed
Anaktisi
This Web-Solution implements a new family of CBIR descriptors. These descriptors combine in one histogram color and texture information and are suitable for accurately retrieving images.
Yes
No
0.225M
University
Open
BRISC
BRISC is a recursive acronym for BRISC Really IS Cool, and is (conveniently enough) also an anagram of Content-Based Image Retrieval System.
Yes
No

University
GPL
Caliph & Emir
Creation and Retrieval of images based on MPEG-7.
Yes
No
Desktop-based
University
GPL
CIRES
developed by the University of Texas at Austin.
Yes
No

University
Closed
FIRE
Open source query by visual example CBIR system. Developed at RWTH Aachen University. FIRE is a research system developed with extensibility in mind and can easily be combined with textual information retrieval systems.
Yes
No

University
Open
GNU Image Finding Tool
Query by example image search system.
Yes
No
Desktop-based
GNU
GPL
ISSBP
Similar Image Search by Imense plugin for Adobe Bridge, free beta.
Yes
Yes
free-beta limited to 4k images
Private Company
Closed
img(Rummager)
Image retrieval Engine (Freeware Application).
Yes
No
Desktop-based
Individual
Closed
imgSeek
photo collection manager and viewer with content-based search and many other features.
Yes
No

Individual
GPL
IKONA
Generic CBIR system - INRIA - IMEDIA
Yes
Yes

University
Closed
MIFile
Image similarity search engine based on MI File (Metric Inverted File).
Yes
No
100M
Research Institute
Open
MUVIS
CBIR System at TUT- Tampere University of Technology.
Yes
No
Desktop-based
University
Closed
PIRIA
CBIR tool developed at CEA-LIST, LIC2M (Multimedia Multilingual Knowledge Engineering Laboratory).
Yes
Yes
3M
University
Closed
PicsLikeThat
Image search using visual similarity search and sorting combined with a recommender system. (Cooperation of pixolution, fotolia and HTW Berlin)
No
No
12M
University
Closed
Pixcavator
Similar image search based on topological image analysis
Yes
No
Desktop-based
Private company
Closed
pixolu
visual & semantic image search engine, by pixolution GmbH & HTW Berlin.
Yes
No

Private Company / University
Closed
RETIN
Interactive images retrieval system - CNRS - ETIS Lab., MIDI Team
No
No

University
Closed
Retrievr
Search and explore in a selection of Flickr images by drawing a rough sketch or uploading an image.
No
No

University
Closed
SIMBA
demo of system by the Albert-Ludwigs-Universitet Freiburg (Germany) Inst. for Pattern Recognition and Image Processing
Yes
No
0.002M
University
Closed
TagProp
The demonstration of image annotation tool TagProp in ICCV2009 for image set: Corel 5k ESP Game IAPR TC-12 and MIR Flickr.
No
Yes

Institute
Closed
VIRaL
Visual Image Retrieval and Localization: A visual search engine that, given a query image, retrieves photos depicting the same object or scene under varying viewpoint or lighting conditions. Using Flickr photos of urban scenes, it automatically estimates where a picture is taken, suggests tags, identifies known landmarks or points of interest, and links to relevant Wikipedia articles. It currently supports 39 cities around the world.
Yes
Yes
2.221M
University
Closed
Windsurf
A general framework for efficiently processing content-based image queries with particular emphasis to the region-based paradigm; it provides an environment where different alternatives of the paradigm can be implemented, allowing such implementations to be compared on a fair basis, from the points of view of both effectiveness and efficiency.
Yes
No

University
Closed



LIMITATIONS OF CONTENT-BASED IMAGE RETRIEVAL

There is a limited results in current CBIR.People ignore lessons about feature selections and the curse of "dimensionality" in pattern recognition because there is little connection between pixel statistics and the human interpretation of an image(the semantic gap). The use of large numbers of generic features makes highly likely that results will not be scalable,that is they will not hold on collections of images other than the ones used during the development of the method.In other words,the transformation from images to features(or other descriptions) is many to one and when the data set is relatively small,but as the size of the set increase unrelated images are likely to be mapped into the same features.


Reference :
http://en.wikipedia.org/wiki/List_of_CBIR_Engines
http://www.ala.org/ala/mgrps/divs/lita/publications/ital/27/1/wan.pdf
encyclopedia of multimedia (Borko Furth) 2006 & 2nd edition 2008
Feb,d,Siu,C,Zhang,H.Eds.(2008).Multimedia information retrieval and management:Technological
fundamentals and applications.Berlin:Springer
http://www.theopavlidis.com/CBIR/Paper B/vers 3.htm